We are studying a number of unconventional myosins and will focus on myosin XVIIIa for this report. Myosin XVIII genes are found in animals from Drosophila to humans and share many structural features with the conventional myosin II class. They have a motor domain with two IQ motifs and a long tail that forms a coiled coil motif to dimerize the heavy chaings. Drosophila has a single gene for myosin XVIII and this gene can be alternatively spliced to give a number of isoforms. We have expressed two motor domain fragments of Drosophila myosin XVIII that differ in their amino-terminal sequences. One isoform contains a PDZ domain at the N-terminus while the other does not. Neither isoform of this myosin has an actin activated MgATPase, and or binds nucleotide. Both bind to actin in an ATP-independent manner with an affinity of around 1 uM. Interestingly, there is a an equilibrium between two states of the myosin, one that is competent to bind actin and one that is not. Optical trapping experiments demonstrate that the myosin can interact with actin, but there is no net displacement of actin resulting from these interactions. We are also studying mammalian myosin XVIIIa. We expressed full length myosin XVIIIa from mouse and showed that it has a long tail similar to that of myosin II. We are examining whether this tail can form filaments. Preliminary studies show that it can weakly bind ATP, but we see no evidence for hydrolysis.